1. Measuring the Effects of Congestion and Request Location on Transit Signal Priority Matt Dorado Erin Qureshi River Hwang Portland State University

2. Outline  Introduction  Objective  Study Design  Test Scenarios  Results  Conclusions/Future Research

3. Introduction  Transit Signal Priority Concept  “Smart bus” knows location and schedule status  Bus communicates priority request to signal  Local controller provides priority

4. Onboard System PCMIA CardControl Head

5. Priority Call Sequence

6. TSP Criteria  Is the bus within Portland city limits?  Is the bus on route?  Are the bus doors closed?  Has the request already been sent?  Is the bus behind schedule ?

7. Signal Timing Green Extension Standard Plan 46 seconds 25.5 seconds39 seconds 19 seconds 39.5 seconds 34.5 seconds32 seconds 27.5 seconds Green extension allows bus to get through the signal

8. Signal Timing Red Truncation 39.5 seconds 34.5 seconds32 seconds 27.5 seconds 29.5 seconds 42 seconds23 seconds 35 seconds Standard Plan Red truncation allows bus to get started through the intersection sooner

9. Objective To evaluate the relationship between Congestion and the distance at which a bus calls for Signal Priority.

10. Modeling Single intersection Intersection of NE 33rd Ave and Sandy Blvd Portland,Oregon

11. Hardware-in-the-Loop Set up showing desktop computer, CID, and 170 Controller

12. Study Design  Software - Vissim, Sychro  TOD Plans – Not used  Approach v/c ratios - 1.1, 0.9, 0.7, 0.5  Volumes - 1640, 1330, 1040, 735 V/hr  Buses - Route 12, 15 min headways, staggered starts

13. Study Design  Data Collection  Three Categories – Travel Time, Waiting Time, Delay  Travel Times – Fixed 1000 ft section  Waiting time  Delay – Buses only

14. Signal Timing Red Truncation 39.5 seconds 34.5 seconds32 seconds 27.5 seconds 29.5 seconds 42 seconds23 seconds 35 seconds Standard Plan

15. Scenarios Bus stop locations – Far and Near Call distance vs v/c Call Distancev/c Ratio 420 feet1.10.90.70.5 350 feet1.10.90.70.5 280 feet1.10.90.70.5

16. Results-Avg Travel Times Table 1 Average Travel Times Detector Length v/cDirection280 ft350 ft420 ft 1.1 East66.764.864.1 West75.676.380.9 0.9 East62.562.261.8 West75.470.173.0 0.7 East61.462.159.3 West70.068.069.3 0.5 East62.560.860.2 West64.964.767.3

17. Results – Farside

18. Results - Nearside

19. Results – Delay Farside

20. Results – Delay Nearside

21. Results - Wait Time Farside

22. Results - Wait Time Nearside

23. Conclusions/Further Research Near side Far side Call button 2 Hour Cycles

24. Acknowledgments  TriMet  Paul Zebell, Willie Rotich, Bill Kloos - City of Portland  Connie Hotard - PTV America  Peter Koonce, Selman Altun - Kittelson and Associates  Dr. Robert Bertini, Dr. Chris Monsere, Erin Qureshi, River Hwang, Neil Byrne, and Office of Academic Affairs - Portland State University

25. Thank You